Acute traumatic injuries result from dust explosions in the workplace to employees and customers. Traditionally these explosions have been associated with the agricultural and mining industries, but they have also occurred in other industries where a combustible dust is handled or generated, as a foundry, textile, or woodworking. Less spectacularly, dust accumulations can lead to fires which are also undesirable. Traditionally through the use of limited testing apparatus which applies to highly specific situations, data has been obtained which assesses the relative hazards of such combustible dust. For the past six years, however, under NIOSH funding, progress has been made in developing a basic understanding of dust combustion and in obtaining fundamental physical data. This effort has been guided by the problems of primary and secondary dust explosions. Two unique combustion facilities have been developed -- a Premixed Combustion Bomb (PTCB) and a Flame Acceleration Tube (FAT). For various dusts, turbulent burning velocity data has been obtained using the former and parameters characterizing layer dust explosions have been obtained using the latter. Such data was not previously available. For the layered case a simple analytical model has ben developed that reproduces the observed results. The three-year research program presented in this proposal has the purpose of deepening and broadening the knowledge concerning combustible dusts and of applying this knowledge to prevent the occurrence of industrial accidents. Using the PTCB, the turbulent dust flame measurements will be continued with more dusts and at higher levels of turbulence. With the Laser Doppler Anemometer, noninvasive measurements concerning the turbulence can be made, and a good analytical model of such a flame will be completed. Measurements will be continued to be made for a layered dust flame using the FAT. Special emphasis will be placed on the accurate determination of the critical parameters in the analytical model and the process of the acceleration to detonation. With regard to the modeling of both of the above, a badly needed piece of information is the rate of combustion of individual dust particles similar to the d2-law for liquid drops. Through the use of a one dimensional isothermal furnace, such information will be obtained. The most destructive combustion process involving dusts is detonation. How this can occur with layered dust is not at present understood, but an existing detonation tube will be used to investigate this phenomena. Basically, this NIOSH program proposes to examine the major aspects of dust combustion with the exception of smoldering. As such, there will be in the United States a central source of information concerning the industrial hazards presented by combustible dusts. This will allow for the convenient interfacing of scientific and accident data, with the attendant interfacing of two very different groups of individuals -- the researchers and the consumers of research data and the training of students in the area of industrial explosion prevention.